1. "Text Overlay"--In the field of video security systems it is known to introduce text into the video signal to provide a real time record of the details of the event which is being monitored. A disadvantage of such prior art systems is that the information is introduced into the video part of the video signal and hence over-writes a part of the video picture which would have otherwise been recorded.
The present invention overcomes this by inserting the text into the non-video part of the video signal
2. "Toggle"--The secret to the success of switching between a plurality of fixed locations is to be able to quickly and easily identify two, or more, viewing locations "on the fly". This is particularly for viewing targets of short term or temporary interest.
In the context of the present invention "on the fly" means to be performed whilst the VSS is in its normal mode of operation and without interrupting the normal operation of the VSS.
Also, in the context of the present invention the feature of "switching back and forth between two (or more) viewing locations" is referred to as "toggling" between the viewing locations.
However, where the operator is interested in two or more specific locations within the viewing region the PTZ camera must be adjusted manually between these locations. This is quite laborious and tedious for the person operating the PTZ camera since it involves adjusting the pan, tilt, zoom and focus setting for the PTZ camera.
We have discovered that this can be overcome by recording the PTZ camera settings corresponding to the desired locations of viewing into a camera position controller and switching between the locations by recalling the settings for each of the viewing locations. The PTZ camera then is controlled by the camera position controller to adjust itself between the desired locations.
"Preset" PTZ camera viewing locations are known and are widely used, however, they require the operator to make several control operations to set each viewing location and at least two control operations to switch from one viewing location to another.
This is particularly inconvenient in a viewing region such as in a casino where there are many potential targets, where there is a requirement to quickly and reliably set viewing locations and where it is critical to be able to switch between typically two targets. The two targets would typically be a suspected felon and a handbag which it is suspected that the felon will steal. Typically, there is an accomplice to be monitored as well. Usually the accomplice in this scenario is a woman who collides briefly with the felon and takes the handbag once the handbag has been stolen. Hence, the task of the operator of the VSS is to first monitor the suspected felon and the handbag by toggling back and forth between the two and once the handbag has been stolen to monitor the felon and then to monitor a person who may collide with the felon.
In the present invention we have determined a way to enable the operator to operate a single control function (such as by a control button) to set, update and toggle between two or more viewing locations. This most reliably and efficiently allows the operator to monitor the targets.
3. "Signal Equalisation"--Prior art video security systems have the problem that equalisation of the video signals must be provided for each video camera. That is, there must be one equalisation unit for each camera.
We have overcome this difficulty by having one equalisation unit for each video monitor and automatically adjusting the equalisation unit for each video camera as it is connected to the monitor. Hence, only a fraction of the number of equalisation units is required.
4. "Rotation End Stops"--Another problem of existing video security systems is that most of the cameras which they use are only able to rotate through about 360.degree.. Hence, when following a target the camera can reach its end stops and have to unwind before it can continue to follow the target. By the time that the camera has unwound the target may no longer be in sight.
We have overcome this problem in an inexpensive way by using a cable loop wound in a way that it is able to twist 3 to 4 times, and hence the camera can be rotated through an angle of about 1000.degree.. A control unit of the camera then is programmed to return the camera to about the centre of its rotation when the camera is not being used to track a target. Hence, the camera is relatively unlikely to reach its end stops during normal use.
5. "Pan Speed vs Zoom"--Prior art vide cameras have the further problem that when the lens is zoomed in (that is viewing an object at a distance) the rate of pan per degree of arc is the same as when the lens is zoomed out. Hence, it becomes difficult to accurately control the pan of the lens when zoomed in. Also, it is difficult to observe the image of the camera during panning when zoomed in because it is usually blurred or indistinct due to the pan speed being too high.
We have overcome this problem by linking the pan speed to the zoom position of the lens so that the rate of pan per degree of arc decreases as the lens is zoomed in and increased as the lens is zoomed out.
6. "Accuracy of Camera Positioning"--Conventional camera position controllers have problems in maintaining accuracy of positioning of the video camera (for viewing a target in a predetermined location) whilst seeking maximum speed of movement of the video camera. That is, in the race for maximum speed, so that one camera can be used to monitor two or more locations, the position controller is not able to keep accurate record of the location of the camera. This is primarily due to problems encountered in rapid changes in acceleration.
We have overcome this problem by designing a camera position controller which ensures that there is a minimum period of substantially zero acceleration before any change in acceleration direction. This thus avoids large changes in momentum introduced when switching abruptly from acceleration to deceleration.
7. "Non-Recording Time"--Another problem of prior art video security systems is that during the times referred to in item 7 of the Field of the Invention no recording can occur. Hence, the video security system can not record all events which its cameras see. For example, in a video security system having 125 VCRs there would be a minimum recording time loss of 2 hours per day due only to the time taken to change over tapes (assuming the operation takes 30 seconds and each tape records for 8 hours). There is further loss of recording when an operator deliberately takes a VCR off-line in order to review one of the tapes.
In order to overcome these periods of lost recording we have designed a VSS to automatically manage the VCRs and cameras so as to ensure that each target is monitored continuously and each video signal is recorded continuously. This is achieved by using backup video cameras and backup VCRs and by providing a log of the camera and VCR associated with each viewed target.